J. Org. C h e m . 1993,58, 904-912
904
Design, Synthesis, and Control of Conducting Polymer Architectures: Structurally Homogeneous Poly(3-alkylthiophenes) Richard D. McCullough,' Renae D. Lowe, Manikandan Jayaraman, and Deborah L.Anderson D e p a r t m e n t of Chemistry, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213-3890 Receiued September 9,1992
The full details of a facile synthesis of structurally homogeneous poly(3-alkylthiophenes) (PAT'S) is presented. In three steps from 3-bromothiophene,PAT'S can be made with completeregiochemical control. We define structurally homogenous as a regiochemically well-defined polymer structure that in our case contains almost exclusively head-to-tail couplings (HT-HT) (2,6-couplings between adjacent thiophene rings). By analysis of the NMR data, our poly(n-butylthiophene) (Sa) contains 93% HT-HT couplings,98% of the desired regiochemistry is found in poly@-hexylthiophene) (6b), 97 % in poly@-octylthiophene),'(Bc),and 96% in poly@-dodecylthiophene),(a). Quenthhgstudies on reactive intermediates and l3C NMR data also demonstrate the regiochelaical purity of these materials. These PAT'S show lower energy absorption maximum ehifta of up to 14 nm in solution, 46 nm in the solid state, and other intense lower energy peaks with shifta of up to 129 nm (609nm) from PAT'S prepared by the usual methods. All of these data are iqdicativeof longer mean conjugation lengths. Molecular mechanics and ab initio calculations were performed on model trimers of 3-nalkylthiophenes and show the relationship between regiochemistry of the trimer and its resultant conformations. The results of these calculations are related to the resultant electrical conductivity in these materials. These poly(3-alkylthiophenes) provide for the first time well-defined structures for the investigation of structure-property relationships in this class of electronic and photonic materials.
Introduction Since the initial discovery of conducting organic polymers, chemistshave sought to prepare polymers that both are soluble (and/or processable)and exhibit high electrical conductivities.' Such an advancement must be made if conducting polymers2 are to fulfill their promise as materials for molecular electronic devices. The early problems concerning material intractability have been thoroughly addressed,3 and now efforts may be focused on the molecular engineering of structures in order to produce materials which possess enhanced electrical and optical properties. An effective design strategy must be aimed at controlling both the microscopic and solid-state macroscopicstructure because they collectivelydefine the resultant band structure* of a given material and thereby determineits electrical6and opticalproperties.6 Molecular engineering must begin with the synthesisof homogeneous (1)Skotheim, T. A., Ed. Handbook of Conducting Polymers; Marcel Dekker: New York, 1986. (2) For some excellent reviews, see: (a) Patil, A. 0.; Heeger, A. J.; Wudl, F.; Chem. Rev. 1988,88,183. (b) Baughman, R. H.; Bredaa, J. L.; Chance, R. R.; Eleenbaumer, R. L.; Shacklette, L. W. Chem. Reo. 1982, 82,209. (c) MacDiarmid, A. G.; Epstein, A. Faraday Diecuss. Chem. SOC. 1989,88,317. (d) Reynolds,J. R. Chemtech 1988,18,440. (e) Kanatzidia, M. G. Chem. Eng. News 1990,68(49), 36. (3) (a) Blankespoor, R.L.; Miller, L. L. J. Chem. Soc., Chem. Commun. 1985,90. (b) Sato, M.; Tanaka, S.;Kaeriynma, K. J. Chem. SOC.,Chem. Commun. 1986,873. (c) Jen,K.-Y.; Miller, G.G.;Elsenbaumer, R. L.J. Chem. Soc., Chem. Commun. 1986,1346. (d) Hotta, S.; Rughooputh, S. D.D. V.; Heeger,A. J.; Wudl,F. Macromolecules 1987,20,212. (e) Nowak, M.; Rughooputh, S.D. D. V.; Hotta, S.;Heeger, A. J. Macromolecules 1987,20,965. (0 Bryce, M. R.; Chiesel, A.; Kathirgamanthan, P.; Parker, D.; Smith, N. R. M. J. Chem. SOC.,Chem. Commun. 1987,466. (g) Ruhe, J.; Ezquerra, T. A.; Wegner, G. Synt. Met. 1989,28, C177. (4) For an excellent introduction to band structures, see: Hoffmann, R. Angew. Chem., Int. Ed. Engl. 1987,26,846. (5) (a) Bredas, J. L.; Street, G.B. Acc. Chem. Res. 1985,18,309. (b) Cowan, D. 0.; Wiygul,F. M. Chem.Eng.News 1986,64(29), 28. (c) Bredas, J. L. J. Chem. Phys. 1985, 82, 3809. (d) Bredas, J. L.; Street, G. B.; Themans, B.; Andre, J. M. J. Chem. Phys. 1986,83,1323. (e) Praneta, J. P.; Grubbe, R. H.; Doqherty, D. A. J. Am. Chem. SOC.1988,110,3430. (0 Lee, Y.-S.; Kertesz, M.; Eleenbaumer, R. L. Chem.Mater. 1990,2,526.
0022-3263/93/196&0904$04.00/0
structure^.^^^ This is the first critical step toward the molecular engineering of organic materials that will undergo controlled macroscopic assembly and poeeeee exceptional properties. A clear indicationof the relationship between structural homogeneity and improved electrical and optical properties is found in Naarman polyacetylene.' Classical synthetic methods produce defective polyacetylene with sp3-hydridizeddefects at junctions of cross-links between polymer chains.@Given that the degree of r overlap along the chain directly determines the band widths (and gap) and sp3 centers decrease the conjugation, classically prepared polyacetyleneswill possess larger band gaps than a more structurally homogeneous material. Indeed, a number of theoretical studies have shownthat coplanarity along the chain r orbitals leads to better electrical propertieskld and nonlinear optical properties.6 AB one example, Naarman's defect-free polyacetylene exhibits electrical conductivities close to that of copper,' in the 12-doped material, and is expected to exhibit large thirdorder susceptibilities.6 It is clear that the ability to design and controlconjugated r architecturesis the key to creating new advanced materials, and some elegant examples have recently appeared in the literature.6J0 (6) (a) Heeger,A. J.; Orenatein, J.; Ulrich, D. R., a. Nonlinear Optical Properties of Polymers. Mat. Res. Soc. Symp. Proc. 1988, 109. (b) Williams, D. J. Angew. Chem., Int. Ed. Engl. 1984,23,690. (c) de Melo, C. P.; Silbey, R. Chem. Phys. Lett. 1988, 140, 537. (d) de Melo, C. P.; Silbey, R. J. Chem. Phys. 1988,88,2567. (e) Ramasenha, 5.; Sow, 2. G. Chem. Phys. Lett. 1988,153,171. (7) Basescu, N.; Liu, LX.;Moses, D.; Heeger, A. J.; Naarmann, H.; Theophilou, N. Nature (London) 1987,327,403. (8) For some additional examples of the synthesis of homogeneow conducting structures, see: (a) Gin, D. G.;Conticello, V. P.; Grubb, R. H. J. Am. Chem. SOC.1992,114,3167. (b) ten Haeve, W.; Wynberg,H.; Havinga, E. E.; Meijer, E. W.J. Am. Chem. Soc. 1991, 113, 6887. (c) McCullough, R. D.; Lowe, R. D. J. Chem. Soc., Chem. Commun. 1992, 70. (9) Chian, J. C. W. Polyacetylene: Chemistry, Physics, and Material Science; Academic Press: New York, 1984.
68 1993 American Chemical Society
Structurally Homogeneous Poly(3dkylthiophenes)
--
J. Org. Chem., Vol. 58, No.4, 1993 901 R
A. FeCI,
HH and olher couplings
*'
\ @-$(y-$ I M" - .
1
Figure 2.
2. 1.m Ni calalvzed
polymerization
'R
R'
A Head-lo-Tail.Head.to-Tail Coupliry (HT.HT)
A Tail-lo-Tail-Head4o-Tail Coupling ( l l - H T )
Head-Io-Tail-Head-to-Head Coupling (HT-HH)
A Tail-lo-Tail-Head-to-Head Coupling ( l l - H H )
Scheme I. Regioselectivity Synthesis of Poly( 3-alkylthiophenes) Br
E)
Poly(3-alkylthiophenes)represent a class of conducting polymers that are soluble and processable and yet retain the magnitude of electrical conductivity of the insoluble parent, polythiophene. Similarto classicalpolyacetylene, the standard synthetic methods (Figure 1)used to make PAT's generate a large number of defecta due to the random couplingdl at the 2,5 positions on the thiophene ring. The resultant structuresmust contain large numbers of thiophene rings that are twisted far out of conjugation due to steric interactions between alkyl chains (a detailed reasoning is given in electronic spectra section). A structurally homogeneoushead-to-tail (HT) arrangement would therefore improve the material's electronic and A synthesis of this regiochemical optical arrangement requires a synthetic method that give absolute regiocontrol at each coupling step in the polymerization reaction. We have very recently developed a synthetic route that allows for complete regiochemical control and produces, for the first time, 8tructurally homogeneous poly(& alkylthiophenes) (PAT'S).'& We define structurally homogeneous as a regiochemically well-defined polymer structure that in our case containsalmost exclusivelyheadto-tail couplings. These poly(3-alkylthiophenes) provide soluble, processable, and well-defined structures for the investigation of structureproperty relationships in this class of materials. Presented herein are the full details of the synthesis of these materials, quenching studies on intermediates in the Synthesis, comparative lH and I3C NMR spectra versus classical PAT's, solution- and solidstate electronic spectra, and molecular mechanics and ab (10)(a) Wudl, P.: Kobayashi, M.; Heeger, A. J. J. O g . Chem. 19f44,49, 3382. (b) Tour, J. M.; Wu, R.; Schumm, J. S.J. Am. C h e m SOC.1990, 112,5662. (c) Schenk,R.;Gregorioue,H.; Meerholz,K.; Heinze, J.; Mullen, K. J. Am. Chem. Soc. 1991,113,2634.(d) Tour, J. M.; Wu, R.; Schu", J. S. J. Am. Chem. SOC.1991,113,7064. (e) Collard, D.M.; Fox, M. A. J. Am. Chem. Soc. 1991, 113,9414. (0Lambert, T.; Ferraris, J. P. J. Chem. SOC.,Chem. Commun. 1991,752. (11) (a) Sato, M.; Morii, H. Polym. Commun. 1991,32,42. (b) Sato, M.; Morii, H. Macromolecules 1991,24,1196. (12)Reeulta by Elsenbaumer have shown that the polymerization of a 3-butylthiophene-3-methylthiophenedimer which contains a 6337 mixture of a HT-HH couplinp leads to a 3-fold increase in electrical conductivity over the random copolymerization reaction; see: Elsenbaumer, R. L.; Jen, K.-Y.; Miller, G. G.; Eckhardt, H.; Shacklette, L. W.; Jow, R. Electronic Roperties of Conjugated Polymers; Kuzmany, H., Mehring, M., Roth, S.,Eds.;Springer Series in Solid State Sciences 1987, 76,400.
1. LDA I THF I -40% I 4 0 min 2. MgBrz.OEtz I -60° to -40% /
40 min
mC HT-HTk Ihe desIrabIs sInrcIure and In the w e where R=d&yI, Ihepmntaoa Of HT-HTCOuplIflgSIS 54%"
Figure 1.
R
4
3. -4OOC + -5OC / 20 min 4. Ni(dppp)Clz / -5' + 25OC I 18 h
R
R
R
5
R Group 8 : n-Butyl b: *Hexyl c: rroayl d: mDcdecyl
Ylelds 69% 36% 65%
33%
initio calculations on model oligomers of PAT's. These studies show the importance of the design, synthesis, and control of r architectures in the preparation of new advanced materials. These data show the differences between the physical properties of these PAT'Sand PAT'S made by standard methods and give new ineighta into the structures of these materials. In addition, preliminary results have shown that these structures exhibit enhanced conductivitieslhversus previously prepared materials and they are expected to show improvements in their nonlinear optical behavior. Results and Discussion
Synthesis of Structurally Homogeneous Poly(3alkylthiophenes). Our synthetic design strategy was to generate homogeneous structures with well-defined regiochemistry by cro~s-coupling~~ target compounds such as 1 (Figure 2). The synthesis that leads to structurally homogeneouspoly(3-alkylthiophenee)is shown in Scheme I. The 3-alkylthiophenes (3) were prepared in 6040% yields.14 Contrary to the regiospecific bromination of 3-methylthiophenewith NBS,the bromination of 3 at the 2-position requires the use of Br2 to give 4, as reported by Gronowitz.15 The polymerization is performed in a oneflask reaction consisting of metalation of 2-bromo-3alkylthiophene4 selectively at the 5-position, followed by trapping the 2-bromo-3-alkyl-5lithiothiophene with magnesium bromide etherate to afford 1. Subsequent treatment of 1, in situ with Ni(dppp)C12,14J6leads to regiochemically defined head-to-tail-coupled poly(3-alkylthiophenes). (13)Tamao, K.; Sumitani, K.; Kumada, M. J. Am. Chem. SOC.1972, 94,9268. (14)(a) Tamao, K.; Kodama, S.;Naajima, I.; Kumada, M.; Minato, A.; Suzuki, K. Tetrahedron 1982,38,3347.(b) Van Pham, C.; Mark, H. B., Jr.; Zimmer, H. Synth. Commun. 1986,16,689.(c) Cunningham, D.D.; Laguren-Davidson, L.; Mark, H. B., Jr.; Van Pham, C.; Zimmer, H. J. Chem. Soc., Chem. Commun. 1987,1021. (15)Consiglio,G.;Gronowitz,S.;Homfeldt, A.-B.;Malteeeon,B.; Noto, R.; Spinelli, D. Chem. Scripta 1977, 11, 175. (16)(a) Yamamoto, T.; Hayashi, Y.; Yamamoto, A. Bull. Chem. SOC. Jpn. 1978,51,2091.(b)Kobayashi, M.; Chen, J.; Moraes, T. C.; Heeger, A. J.; Wudl, F. Synth. Met. 1984,9, 77.
906 J. Org. Chem., Vol. 58, No.4, 1993
We have applied this method in the preparation of four different alkyl-substituted polythiophenes and a few polyether-substituted p01ythiophenes.l~We expect that (providingthat R is compatiblewith the coupling catalyst) this route should be general for the synthesis of a large number of poly(3-substituted thiophenes) of well-defined structure. One of the key features of our synthesis is the ability to metallate 4 with LDA in the 5-position18both regioepecificallyand without extensivescramblingof the generated organolithium. In previous studies on thienyllithium species, we and others have found that regiospecific metalation of brom~thiopheneal~ and 3-alkyl thiopheneal7B occurs either at the 2- or 5-positionand not at the 4-position on thiophene. However, under certain conditions, quenching studies with water or a variety of other carbon nucleophiles have shown that scrambling may This positional exchange of the lithium and halogen on a single thiophene ring is well established and is known as the "halogen dance" mechanism. We have performed quenching studies to examine whether the 2-bromo-3alkyl-5-1ithiothi0phene~~ generatedunder these conditions undergoes scrambling and find little to no scrambling occurs. In this case, scrambling can occur by a number of pathways, including the "halogen dance" mechanism and/or a bimolecular metal-halogen exchange between the generated lithium species and the starting material 4. If the "halogen dance" mechanism occurs,quenchingwith TMSCl should yield 2-(trimethylsilyl)-3-butyl-5-bromothiophene and/or 2-bromo-3-alkyl-4-(trimethylsilyl)thiophene. A bimolecularmetal-halogen exchange would lead to 2-(trimethylsilyl)-3-alkylthiopheneand 2,5-dibromothiophene as first-generationproducts. We have found that quenching with TMSCl after treatment of 4 with LDA and also after generation of the Grignard reagent (step 2 in 4 5) provides C1-3% of scrambled product. In the case of R = n-butyl, we have trapped the lithio species and recover only 99% of 2-bromo-3-butyl-5(trimethylsily1)thiopheneand C1% of 2-(trimethylsilyl)3-butyl-5-bromothiopheneas determined by 'H NMR and GC/MS. In the case of 2-bromo-3-dodecylthiophene, quenchingboth the Grignard and the lithio derivativeand 97-98?6 of 2-bromo-3-dodecyl-5-(trimethylsilyl)thiophene waa obtained according to NMR and GUMS and 2-3 % of the scrambled product by NMR. The same results are observed in quenching studies on the 3-hexylthiophene derivatives. The quenching experimentsindicate that the scrambling is due to a "halogen dance" mechanism. In conclusion, the lack of scrambling is entirely consistent with the structural homogeneity of the poly(3-alkylthiophenes) 5, and their observed lH and 13CNMR spectra.
-
(17) Lowe, R. D.; Williams, S. P.; McCullough, R. D. Unpublished results. (18) (a) Soucy-Breau, C.; MacEachern, A.; Leitch, L. C.; Arnason, T.; Morand, P. J. Heterocycl. Chem. 1991,28, 411. (b) MacEarchern, A.; Soucy, C.; Leitch, L. C.; Arnason, T.; Morand, P. Tetrahedron 1988,44,
2403. (19) (a) van Phan, C.; Macomber, R. S.; Mark, H. B., Jr.; Zimmer, H. J.Org. Chem. 19&1,49,5260. (b) Daviee,G. M.;Daview,P. S. Tetrahedron Lett. 1972, 8507. (c) Kano, S.; Yuasa, Y.; Yokomatsu, T.; Shibuya, S. Heterocycles 1983,20, 2035. (20) Krische, B.; Hellberg, J.; Lilja, C. J. Chem. SOC.,Chem. Commun. 1987,1476. (21) For the sake of clarity in presentation, we have taken the liberty
of naming our compounds with 3-alkylgroup always remaining at position 3 and numbering other substituents accordingly.
McCullough et al. a
I
I
I
I
I
I
I
I
9.0
6.0
7.0
6.0
5.0
4.0
3.0
I 2.0
I
I
1.0
0.0
PPM
i I
I
I
I
I
180
160
140
120
100
I 80
I
I
I
60
40
20
PPM
Figure 3. Full lHand 13CNMR of PHT prepared by the present method.
NMR Spectroscopy In 1990, Sato and Moriill deduced through a series of two-dimensional 'H NMR studies that 54% of electrochemically synthesized poly(3-dodecylthiophene)consieted of head-to-tail head-to-tail couplings. Therefore,almost half of the couplingsin poly(3-alkylthiophenea)are defects with regard to the conducting polymer architecture. In their analysis, the four singlets in the aromatic region can be clearly attributed to the protons on the 4-position on the thiophene ring where each peak results from a different type of trimeric sequence of HT-HT (6 = 6.98), HT-HH (6 = 7.00), TT-HT (6 7.02), and TT-HH (6 7.05) linked thiophenerings. In contrast, our synthesisof PAT'S afford strikingly clear 'H and 13CNMR spectra (Figure 3). As an example, Figures 3 and 4 show the full proton and carbon spectrum and the expanded aromatic and methylene regions of the lH NMR spectra of poly(& hexylthiophene) (PHT) as synthesized by this method. As a comparison the expanded lH NMR for PHT as prepared by the FeCbmethod is shown in Figure 4. Figure 5 shows the l3C expanded aromatic region for our PHT. Both PHT's were prepared in our laboratory. The 'Hand 13C NMR spectra,11v22123 of a PAT as synthesized either electrochemicallyor chemically or by other methods have been published as part of other studies on the polymerization of dialkybithiophenes and related work and show a large number of imperfect couplings and random (22) (a) Elsenbaumer, R. L.; Jen, K.-Y.; Oboodi, R. Synth. Met. 1986, 15, 169. (b) Zegorska, M.; Krische, B. Polymer 1990, 31, 1379.
J. Org. Chem., Vol. 58, NO.4, 1993 907
Structurally Homogeneous Poly(3-alkylthiophenes)
I
b
I
I
I
I
I
I
I
I
I
7.08
7.08
7.M
7.02
7.00
6.88
6.96
6.94
6.92
I
series show exclusively four predominant absorptions attributable to the carbons on one regiochemicallydefined thiophene ring. Absorptions at 128.5, 130.5, 134.0, and 140.0 represent the head-tu-tail-coupledpoly(3-hexylthiophene) structure.23 A 13C NMR of PHT made from FeC4, in addition to the HT-HT couplings,exhibits welldefined peaks at 125.2, 126.6, 127.4, 128.3, 129.6, 134.9, 135.7,136.8,140.3,142.9,and 143.4 ppm. These absorptions represent the carbons associated with non-head-totail regiochemical isomers. Poly(3-hexylthiophene),prepared by our method, shows only very small absorptions at 127.4 and 126.7 and peaks barely distinguishable from the base line at 140.3 and 137.0, which correspond to the 1-2 % of scrambling in the 2-bromo-3-alkyl-5-lithiothiophene starting material. Based on these data and other literature l3C NMR data,23a we find no detectable headto-head-tail-to-tail couplingin our alkylthiopheneunits.24 These NMR results are in stark contrast to previously prepared materials.
PPM
Conformational Behavior of Oligomerr of Alkylthiophenes
I
l 7.08
l
7.08
I
7.04
I
I
I
I
I
I
7.02
7.00
6.88
6.00
6.94
6.92
f
PPM
Figure 4. (a) Expanded lH of PHT aromatic region prepared by this method. The inset shows the expanded lH NMR of the PHT aryl methylene region. (b) Expanded lH NMR of PHT aromatic region prepared by the FeC13 method. The inset is the aryl methylene region. Both of these PHT samples are treated identically.
regiospecificity. The IH and 13CNMR spectra show that our polymers contain almost exclusively head-to-tail couplings. By analysis of the NMR data, our PHT contains approximately 987% HT-HT couplings, 1% HT-HH couplings, and 17% TT-HT couplings. A similar analysis indicates 937% of the desired regiochemistry in our PBT, 95 7% in PDDT, and 97 7% in POT. These data show that we have been able to gain regiocontrol over the polymerization reaction, which leads to essentially one structure. The 13CNMR reinforces the interpretation of the 1H NMR data and shows the very high structural regularity in polymers prepared by this method. Figure 3 shows the full 13C spectrum for our poly(3-hexylthiophene). The alkyl region only shows six absorptions corresponding to the hexyl side chain at 14.9,23.6,29.8,30.3,30.9, and 32.0 ppm. Shown in Figure 5 is the expanded 13C aromatic region for PHT. The spectra for PHT and the entire PAT
The effect of a microstructural irregularity is to create a sterically driven twist of the thiophene rings out of coplanarity and conjugation with one another. Thie is illustrated by the structural diagrams in Figure 6. The larger the torsion angle between thiophene rings, the greater the bandgaps will be in typical poly(3-alkylthiophenes) as compared to a structure with well-defined head-to-tail regiochemistry. Since the high electrical conductivity in the oxidized polymer results from the the bandgap development of mid-gap bipolaron (and width) of the neutral polymer directly determines the electrical conductivity in these materials. It is currently accepted that the mechanism for conductivity in the polythiophenes, called the bipolaron mechanism, proceeds through a combination of a charge movement along the polymer chain25126(where charge carriers are bound dications) and through a polymer-to-polymer (and grain to grain) charge-hopping mechani~m.~'The energetic costs of moving charge along a conjugated chain is directlyrelated to the amount of molecular orbital overlap. It is, therefore, absolutely critical to have structurally homogeneous conducting polymers in order to allow for (23)(a) Souto Maior, R. M.; Hinkelmann, K.; Eckert, H.; Wudl, F. Macromolecules 1990,23,1268.Wudl's 1% of the PHT prepared by the FeC13 method w a ~used as a guide for the 13C assignment of the regiochemically defined material, poly(3,3-dihexyl-2,2-bithiophene). (b) Hotta, S.; Soga, M.; Sonoda, N. Synth. Met. 1988,219,267.(c) Goedl, W. A.; Somanathan, N. S.; Enkelmann, V.; Wegner, G. Makromol. Chem. 1992, 193, 1195. It should be noted that most of the PAT'S in these references have been precipitated after isolation-a method developed by Wudl. This appears to selectivelyimprovethe ratio of HT-HTcoupled polymer by precipitating the most crystallimepolymers. The improvement appears to be an increaseof 1045% in the percentage of HT-HT polymer. Our polymers have not been precipitated. We have taken carbon spectra of PHT prepared by the FeC& method and treated identically to ow PHT and have obtained spectra inferior to the spectra found in these references. (24)Since very smallamounta of monomers posaeestheregiochemistry that would lead to non-head-to-tail coupling,the probability of two 'bad" couplings occurring in a row is remote. (25)Very interesting model studies have recently been reported, see: (a) Tolbert, L. M.; Ogle, M. E. J. Am. Cham. SOC. 1990,112,9519.(b) 1991, Caper, J. V.; Ramamurthy, V.; Corbin, D. R. J. Am. Chem. SOC. 113,600. (c) Hill, M. G.; Mann, K. R.; Miller, L. L.; Penneau, J.-F. J. Am. Chem. SOC.1992,114,2728. (26)Child, A. D.;Reynolds, J. R.J.Chem.SOC.,Chem. Commun. 1991, 1779. (27) Jeon, D.;Kim, J.; Gallagher, M. C.; Willis, R. F. Science (Washington, D.C.) 1992,256,1662.
908 J. Org. Chem Vol. 58, No.4, 1993
145.0
140.0
135.0
130.0
McCullough et d.
125.0
PPM
Figure 5. Expanded 13C NMR of PHTaromaticregion prepared by this method.
less mnjugation or overlap smaller Asmaller bandwahs larger bandgws lower conductivities
Figure 6. maximum intramolecular orbital overlap (large transfer integrals) as well as interchain stacking and overlap to reduce highly resistivecharge transport pathways. Given that our poly(3-alkylthiophenes)have very few undesirable couplings, adjacent thiophene rings can now access a coplanar (or nearly coplanar) conformation of low energy. In order to investigate the energetic consequences of ring coplanarity in poly(3-alkylthiophenes)with desirable HT couplings compared to the undesirable HH couplings, we have examined the conformations of model dimers, trimers, and tetramers of alkylthiophenes by both mo-
lecular mechanics and ab initio calculations.M.28 Resulta from the molecular mechanics calculations indicate that the lowest energy conformations for structures like A (Figure 6) have all the rings nearly coplanar. In contrast, structures like B have lowest energy conformationsthat possess a large twisting of the rings out of coplanarity at the HH junction (torsional angle between ring 7 and 8, in B). A detailed discussion of these results will appear in a forthcoming article; below we describe our preliminary findings.29 Molecular mechanics performed on a 3-butylthiophene trimer with HT-HT coupling yield lowest energy conformations in which the thiophene rings are trans to each other and within 20' of coplanarity. The potential energy surfacefor twisting the rings from a coplanarconformation to -20' out of coplanarity is very flat, the structures in this region are within 5 kcal/mol higher in energy. Although these are gas-phase calculations and packing forces to a great extent determine the solid state structure, it is insightful to see that a head-to-head coupling, in a models for poly(3-alkylthiophenes),causesthe torsional angles to be approximately 40' out of coplanarity. In contrast, a head-to-tail coupling can access very low energy structures that can possess coplanaritybetween alkylthiophenerings and lead to greater along the chain ?r overlap and small bandgaps. Solution- and Solid-state Electronic Spectra A qualitative measure of ?r orbital overlap can be observed in the electronicspectra of conjugated molecules. In conjugatedpolymers the extent of conjugation directly (28) (a) Barone, V.; Lelj, F.; R w o , N.; Toscano, M. J. Chem. SOC., Perkin Trana.2 1986,907. (b)Bredae,J. L.; Heeger,A. J. Macromolecules 1990,23, 1150. (29) McCullough, R. D.; Brooks, C. L., 111; White, C. A. Manuscript
in preparation.
J. Org. Chem., Vol. 58, No.4, 1993 909
Structurally Homogeneous Poly(3-alkylthiophenes) Table I. A Comparison of Solution- and Solid-state UV-vis Data of Poly(3-alkylthiophenes) soln aoln solid-state solid-state' structure
PBT PHT POT PDDT
Amax
Xmax
(FeCW
(this method)
(FeC13)*
436 436 436 436
438 442
480
446 450
Xmax
Xmax
480
(this method) 500
480 480
.I6
I
1
I
1
I
I
I
504 520 526
0 The thin f ilm data is independent of the film thickness in the thickness regime of the films used in the measurement (3-10 pm). Cast from CHC13.
affects the observed energy of the r-r* transition, which appears as the maximum absorption in these materials. Experimentaldeterminationof the electronicspectra gives insight to the extent of conjugation and structure of PAT's. The results of both the solid-state and solution UV-vis are shown below in Table I, and the solid-state spectra are shown in Figure 7. A number of interesting trends are observed by inspection. In the solution data, we observe for the first time a change in the Am, as a function of alkyl side chain. As indicated by lower energy shifts in the ,A,, longer conjugation lengths are observed in the PAT's possessing the longer alkyl side chains. Photoelectron spectroscopyand X-ray studies have indicated that PDDT is more planar and stacked (lamellar-like)3Oin structure and PBT is probably more helica131in nature. The solution UV-vis data presented here corroboratea helical structure for PBT and alamellar structure for PDDT. X-ray studies by Winokur and co-workers have further established the lamellar-like structure in PAT's with longer side chains in the solid state.32 It would appear that even in solution an aggregation of the alkyl side chain occurs in PDDT giving rise to greater conjugation lengths versus that of PBT. Preliminary light scattering in solution33results showsuch aggregationof two polymer chains in POT. This aggregation occurs to a great extent at low temperature and leads to a large thermochromic effect,w marked by a deep red solution at 25 OC and a dark purple solution at -40 OC. The shift to lower energy of the A,, in PDDT prepared by our method is found to be 14 nm (450 nm) in solution, 46 nm (526 nm) in the solid state,and other intense lower energy peaks with shifts of up to 129nm (609 nm) relative to PDDT prepared with FeC13. We interpret this effect as an increasing steric interaction between longer alkyl chainsis found in the FeCl3 samples-these PAT's contain undesirable couplings. As noted in Figure 6 in structure B, steric repulsions between the alkyl chain on the thiophene ring 6 and the alkyl chain on thiophene 8 will be at a maximum when the alkyl chain is the longest, thereby decreasing conjugation between thiophene rings. (30) (e) Winokur, M. J.; Spiegel, D.; Kim, Y. H.; Hotta, S.; Heeger, A. J. Synth. Met. 1989,28, C419. (b)Mardalen, J.; Samueleon,E.J.; Gautun, 0.R.; Carlsen, Per H. Solid State Commun. 1991,77,337. (c)Tashiro, K.; Ono, K.; Minagawas, Y.; Kobayashi, K.; Kawai, T.; Yoshino, K. J.
Polymer Sci. (Polymer Physics) 1991, 29, 1223. (31) (a) Garnier,F.;Tourillon,G.;Barraud,J. Y.;Dexpert,H.J.Mater. Sci. 19G, 20, 2687. (b) Yang, R.; Evans, D. F.; Christensen, L.; Hendrickson, W. A. J. Phys. Chem. 1990,94,6117. (32) (a) Winokur, M. J.; Waauley, P.; Modton, J.; Smith, P.; Heeger, A. J. Macromolecules 1991,24, 3812. (b) Prosa, T. J.; Winokur, M. J.; Moulton, J.; Smith, P.; Heeger, A. J. Macromolecules 1992, 25, 4364. (33) Yue, S.; Berry, G. C.; McCullough, R. D. Unpublished resulta. (34) (a) Ycahino, K.; Hayashi, S.; Sugimoto, R. Jpn. J. Appl. Phys. 1986,23, L899. (b) Rughooputh, S. D. D. V.; Hotta, S.; Heeger, A. J.; Wudl, F. J. Polym. Sci., Polym. Phys. Ed. 1987,25, 1071. (c) Inganas, 0.; Salaneck,W . R.; Osterholm, J. E.; Laakso,J. Synth. Met. 1988,22, 395.
Poly (3-dodecylthiophene) .02 0 300
I
I
I
350
400
450
1 500
I
I
1
I
550
600
650
700
750
I
-
Wavelength (nm) L
I
I
I
I
I
I
Poly (3-octylthiophene) .Ol
o c 300
I
I
I
I
I
I
I
4
350
400
450
500
550
600
650
700
I
I
1
Wavelength (nm) .I4
.10 .12
1
oc 300
I
I
I
I
I
py
C
I
I
I
1
I
1
I
350
400
450
500
550
600
650
-I 700
Wavelength (nm)
Figure 7. UV-vis of these PAT's in the solid state. Labeled absorbances are the A,, and fine structure peaks. Label at 675 nm is a marker near the band edge (=Id eV).
Accordingly,the largest should occur in the PAT's with longest alkyl chains. Molecular modeling studies indicate that butyl side chains are long enough to sterically interfere, and therefore increase steric interactions should occur with the longer side chains. A direct comparison of our data with data on highly ordered,very thin films of poly(3-methylthiophene)" gives insight on the macroscopic order in PAT f i b reported here. The highest value for the electrical conductivityfor a PAT is 2000 S cm-l and was observed in 0.2-pm thin films of poly(3-methylthiophene)(PMT),which had a A, of 510 nm. These highly conducting PMT films exhibited (35) AX,, = I(X,,(PAT by our method)- X,.,(PAT byFeChmethod)(. (36) (a) Roncali, J.; Yassar, A.; Garnier, F. J . Chem. Soc., Chem.
Commun. 1988,581. (b) Yassar, A.; Roncali, J.; Garnier, F. Macromol-
ecules 1989, 22, 804.
910 J. Org. Chem., Vol. 58, No.4, 1993 Table 11. Fine Structure in Solid-state UV-vis Data of Structurally Homogeneous Poly(3-alkylthiophenes) solid-state solid-state A,, Amax-relintensity structure (FeCl3)a (our method) PBT 480" 5004 590 580 610 PHT 480b ~504~-2.2 550-1.8 600-1.0 POT 480b 520'-1.6 553-1.6 603-1.0 PDDT 480b 526'-1.8 561-1.6 609-1.0 Broad absorbance exhibiting barely discernable shoulders; relative intensity of absorbances were not determined. Relative intensity is a measure of the absorbancerelativeto the smallestpeak reported. b Broad absorbance with no discernable shoulders. Broad absorbance with well-defined shoulders.
thickness-dependent solid-state UV-vis spectra, which correlated with the conjugation lengths and electrical conductivity. The thinnest films in which conductivity measurements were reported were 0.19 pm, although ,A, values as high 552 nm were observed in the 0.006-clm films. These thin films of PMT had a very high degree of structural order and extended ?r conjugation lengths. In much thicker films (3-10 pm) of our PAT's, we find that the Am, of POT is 520 nm and PDDT is 526 nm. We observe conjugation lengths of the order of very highly ordered poly(3-methylthiophene). These results indicate that the packing arrangements and molecular structures these PAT's are ordered at a comparable level to ultrathin films of poly(3-methylthiophene). In addition, X-ray studieson thin films of our PAT's (in our laboratory) have shown highly ordered structures which are more oriented than stretch-oriented PAT's (from the FeC4 method).37 In the solid-state data, we have also observed a series of well-defined long wavelength shoulders in the UV-vis data for the structurally homogeneous PAT's (Figure 7 and Table 11). Our interpretation is that additional longrange ordered structures exist that give rise to longer conjugation lengths. The "shouldern peaks present are quite substantial as indicated by the relative intensities of the "shoulder" to the Am=. This type of behavior has also been observedfor oligomers of thiophene.% We expect that this relative intensity ratio is an indication of t h e amount of highly ordered phases present in the solid-state thin films.
Conclusion We have presented a synthetic methodology that produces structurally homogeneous poly(3-alkylthiophenes). Due to their structural homogeneity, these materials give a clearer picture of structure-property relationships in conjugated polymers. The ability to acquire data which are unobscured by structurally defective materials is crucial in defining the chemistry and physics of conducting organic materials and the systematics of their nonlinear optical behavior. We believe that the synthesis of well-defined materials is the first step (37) McCullough, R. D.;Tristram-Nagle,S. Manuscript in preparation. (38) Xu, Z.;Fichou,D.;Horowitz,G.;Garnier, F.J. Electroanal. Chem. 1989,267, 339.
McCullough et al. toward the design and engineering of conjugated
T
architectures that will self-assemble in three dimensions and possess exceptional electronic and optical properties.
Experimental Section All reactions were performed under prepurified nitrogen or argon, using dry glassware. Glassware was either dried in an oven overnight or flame dried and then cooled under a stream of argon or nitrogen. Tetrahydrofuran and diethyl ether were dried over Na benzophenone ketyl radical and freshly distilled prior to use. Trimethylsilyl chloride was freshly distilled prior to use. Diisopropylaminewas dried over CaH2 and distilled prior to use. Acetic acid was used as received. All 2-bromo-3alkylthiophenes were freshly distilled prior to use. Both MgBrZaOEt and Ni(dppp)Clz were purchased from the Aldrich Chemical Co. and were used as received and were ideally stored in a drybox. lH and l3C NMR spectra were recorded on an IBM Bruker FT300 spectrometer. All 300-MHz lH NMR spectra were recorded in CDC13and are reported in ppm as S relative to internal tetramethylsilane at 0.0 ppm. The amount of head-to-tail coupling was determined by NMR integration of the small peaks near 7.0 using the Sato and Morii analysis and the peaks near 2.5 using the method of Elsenbaumer et al.l* Both methods gave consistent results. The integrated numbers of protons for the regiochemical isomers other than the head-tu-tail (HT-HT) ones in the lH NMR spectra is given relative to the integrated area of the 4-proton on the major isomer. For example, 2% of a HTHH polymer will be listed as 7.04 (e, 0.02 H), where the HT-HT polymer is 7.00 (a, 1H). All 75.4 MHz 13CNMR were taken in CDC13 and are reported relative to internal, residual CHCl3 at 77.0 ppm. All UV-vis spectra were taken on either polymer solutions in CHCb or polymer thin films cast onto quartz cuvettes using a Hewlett-Packard 8451A diode-array spectrophotometer and are reported in X and are in nm. Infrared spectra were obtained on polymer thin films cast from CHCl3onto NaCl plates using a Nicolet 7199 FTIR and are reported in Y as cm-l. All 3dkylthiophenes and 2-bromo-3-alkylthiopheneswere purified by vacuum fractional distillation and were characterized by lH NMR, GC/MS, and elemental analysis. Gas chromatography/ mass spectroscopy was performed on a Hewlett-Packard 59970 GC/MS workstation. The GC column was a Hewlett-Packard fused silica capillary column cross-linkedwith 5%phenyl methyl silicone. Elemental analysiswas performed by Midwest Microlab, Indianapolis, IN. All molecular mechanics calculations were performed using Charmm software (Polygen, Inc.) on a Silicon Graphics Iris 4D Computer. The 3-butylthiophene trimer structures were optimized at the ab initio Hartree-Fock STO3G level using Gaussian 92.39 Preparation of 3-n-Alkylthiophenes 3a-d. All 3-alkylthiophenes were prepared by the literature procedure of Kumadal" using the specific synthetic details outlined by Zimmer et al.lrb All compounds were purified by distillation and purities checked and structures characterized by lH NMR and GC/MS. All are literature compounds. Zimmer has synthesized and reported data for 3-butylthiophene and 3-hexylthiophene. 3-n-Octylthiophene (3c). The crude product obtained from 12.7 g (78 mmol) of 3-bromothiophene was distilled (79 "C (0.35 mmHg)) to give 9.4 g (61% ) of the product: 'H NMR (CDCl3) 7.23 (dd, J = 4.5 Hz,1 H), 6.93 (m, 2 H),2.62 (t, J = 7.5 Hz,2 H), 1.63 (m, 2 H), 1.38-1.20 (m, 10 H), 0.88 (t,J = 5.7 Hz); MS m/z (relative intensity) 196 (16, M+), 97 (100, CsHaS+). Purity was determined to be >99% pure by GC. 3-n-Dodecylthiophene (3d). The crude product obtained from 16.3 g (100 mmol) of 3-bromothiophene was distilled (175 "C (1.8 mmHg)) to give 18.3 g (73%) of the product: lH NMR (CDC13) 7.17 (dd, J = 2.4, 4.0 Hz, 1H), 6.86 (m, 2 H), 2.54 (t,J = 7 Hz, 2 H), 1.61 (m, 2 H), 1.3-1.2 (m, 18 H), 0.86 (t, J = 6 Hz); (39) Gaussian 92, Revision A Frisch, M. J.; Trucks, G . W.; HeadGordon, M.; Gill, P. M. W.; Wong, M. W.; Foreeman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb, M. A.; Replogle, E. S.; Gomperta, R.; Andrea, J. L.; Raghavachari, K.; Binkley, J. S.; Gonzalez, C.;Martin, R. L.; Fox, D. J.; Defrees, D. J.; Baker, J.; Stewart, J. J. P.; Pople, J. A., Gaussian, Inc., Pittsburgh, PA, 1992.
Structurally Homogeneous Poly(3-alkyithiophenes)
MS m/z (relative intensity) 252 (10, M+),97 (100,C5H@+).Purity was determined to be >99% pure by GC. Preparation of 2-Bromo-3-n-alkylthiophenes 4a-d. All 2-bromo-3-alkylthiopheneswere prepared accordingto Gronowitz et al.15 The boiling point and NMR characterization of 2-bromo3-hexylthiophenehas been reported.l5 2-Bromo-3-n-butylthiophene (4a). Into a dry round-bottom flask was placed 68.5 mL (0.8 M) of acetic acid, which was then sparged with argon (5 min). Then 7.7 g (0.055 mol) of freshly distilled 3-n-butylthiophene (3a) was added. The mixture was cooled to 10 "C, whereupon a 2.5 M solution of bromine (2.8 mL, 0.055 mol) in acetic acid was added dropwise from an addition funnel over a period of 30 min, while the temperature was maintained at 10-15 OC. The material was then stirred in an ice bath for 30 min and was then poured onto ice. The mixture was then extracted into CHCl3, the CHC13 layer washed with NaOH until pH -6 and dried over MgSO,, and the solvent removed by rotary evaporation. The product was twice distilled (80"C (1.8 mmHg)) to yield 5.8 g (48%) of 4a: lH (CDC13) 7.11 (d, J = 5.7 Hz, 1H), 6.72 (d, J = 5.7 Hz, 1H), 2.50 (t,J = 7.4 Hz, 2 H), 1.49 (pentet, J = 7.1 Hz, 2 H), 1.22 (sextet, J = 7.1 Hz), 0.86 (t,J = 7.5 Hz); MS m/z (relative intensity) 220 (17.5, M + 2), 218 (18, M+), 97 (100, C5HsS+). Anal. Calcd for C&BrS: C, 43.85; H, 5.06. Found C, 43.72; H, 5.12. Purity was also checked by GCl MS. 2-Bromo-3-a-hexylthiophene (4b). Compound 4b was prepared in an identical fashion as 4a, except that the crude product was isolated by extraction with E b 0 . The crude product was obtained from a reaction starting with 20.6 g (83 mmol) of 3b. The product was distilled (117 "C (3.8 mmHg)) to give 10.0 g (49%)of 4b: 'H NMR (CDCl3) 7.18 (d, J = 5.5 Hz, 1 H), 6.79 (d, J = 5.5 Hz, 1 H), 2.62 (t, J = 7.5 Hz, 2 H), 1.62-1.52 (m, 2 H), 1.35-1.23 (m, 6 H), 0.89 (t, J = 5 Hz, 3 H); MS mlz (relative intensity) 248 (56, M + 2), 246 (57, M+),97 (100, CsH&+). Anal. Calcd for C1~H15BrS:C, 48.59; H, 6.12; Br, 32.32; S, 12.97. Found C, 48.70; H, 6.17; Br, 32.16; S, 12.99. Purity was also checked by GCIMS. 2-Bromo-3-a-octylthiophene (4c). Compound 4c was prepared in an identical fashion as 4b. The crude product was obtained from reaction starting with 17.07 g (87 mmol) of 3c. The product was distilled (123 "C (1.1mmHg)) to give 9.4 g (40%)of 4 ~ 'H : NMR (CDC13) 7.18 (d, J = 5.7 Hz, 1 H), 6.79 (d, J = 5.7 Hz, 1 H), 2.56 (t, J = 7.5 Hz, 2 H), 1.58-1.50 (m, 2 H), 1.34-1.21 (m, 10 H), 0.89 (t, J = 4.5 Hz, 3 H); MS m/z (relative intensity) 276 (12, M + 2), 274 (13, M+),97 (100, CsHsS+). Anal. Calcd for C12H19BrS: C, 52.36; H, 6.96. Found: C, 52.43; H, 7.09. Purity was also checked by GCIMS. 2-Bromo-3-n-dodecylthiophene (4d). Compound 4d was prepared in an identical fashion as 4b. A preparation of 4d from 9.4 g (0.037 mol) of freshly distilled 3-dodecylthiopheneyielded after distillation (165 OC (0.22 mmHg)) 5.0 g (41%) of product: 'H NMR (CDC13) 7.19 (d, J = 5.3 Hz, 1 H), 6.79 (d, J = 5.3 Hz, 1H), 2.58 (t,J = 7.5 Hz, 2 H), 1.64-1.57 (m, 2 H), 1.36-1.20 (m, 18 H), 0.88 (t,J = 5.0 Hz, 3 H); MS mlz (relative intensity) 332 (3, M+),97 (100, CsHsS+). Anal. Calcd for CleH27BrS: C, 58.00; H, 8.21. Found: C, 58.20; H, 8.25. Purity was also checked by GUMS. General Synthesis of Poly(3-alkylthiophenes) (PAT). All PATS were prepared in procedures identical to the procedure below. In some cases, after the addition of 0.5 mol % of Ni(dppp)C12and stirring for approximately 12 h, another 0.5% of Ni(dppp)Clzwas added. In the case of POT, a double addition of the catalyst resulted in an increase in yield to 65 % from 36 % and an average molecular weight increase from M, = 12 OOO to M, = 24 OOO (by GPC). We have also precipitated the PAT into hexane instead of methanol with no apparent changes in yield or purity. Example preparations are given below. Preparation of Poly(3-hexy1)thiophene Using 0.5 mol % Catalyst. Into a dry round-bottom flask was placed dry diisopropylamine (2.11 mL, 15 mmol) and freshly distilled, dry THF (75 mL, 0.2 M). To the mixture was added 6.0 mL of 2.5 M n-butyllithium (15 mmol) at room temperature. The mixture wascooledto-40OC andstirredfor4Omin. Thereactionmixture containing LDA was then cooled to -78 OC, and 2-bromo-3hexylthiophene (3.7 g, 15 mmol) was added. The mixture was stirred for 40 min at -40 "C. The mixture was then cooled to -60
J. Org. Chem., Vol. 58, No.4,1993 911 OC, M g B r 2 ~ E b 0(3.87 ~ g, 15 mmol) was added, and the reaction was stirred at -60 "C for 20 min. The reaction was then warmed to -40 OC and stirred for 15 min. The reaction was then allowed to slowly warm to -5 OC, whereupon all of the MgBrleEhO had reacted. At -5 "C, 0.48 mol % of Ni(dppp)C& (39 mg, 0.072 mmol) was added. The mixture was allowed to warm to room temperature overnight (-12-18 h). The polymer was then precipitated with MeOH (300 mL), and the resulting red precipitate was then filtered and washed with MeOH, H20, and MeOH again. The solid was then dried under vacuum. Removal of oligomers and impurities was achieved by subjecting the solid to Soxhlet extractions with MeOH first followed by hexanes. The polymer was then dissolved in CHCl3 using a Soxhlet extractor, the CHCl3 was removed, and the residue was dried under vacuum to yield 760 mg of 95 % head-to-tail coupled poly(Bhexylthiophene) (36% yield). The moat recent preparation (PHTprecipitatedwith hexane) resulted in 98% HT-HT-coupled P H T 'H NMR (CDCl3)6.98 (8,l H), 2.78 (t, 2 H), 1.62 (q, 2 H), 1.48 (m, 2 H), 1.36 (m, 4 H), 0.90 (t,3 H); peaks also present that are representative of 5% of other regiochemical isomers are assigned at 7.04 (a, 0.01 H), 7.00 (e, 0.04 H), 2.60 (m, 0.04 H); I3C NMR ('H decoupled) 140.0, 134.0, 130.5, 128.5, 31.8, 30.5, 29.5, 29.3,22.8,14.0; relative intensities are coupled to relaxationtimes; however, the relative intensity of observable small peaks (relative to the 128.5absorption) are 127.4(0.05),126.7 (0.02), 140.3 (0.01), 137.0 (0.01); UV-vis (CDCl3) A,, 442 nm; band edge 550 nm. (Film, cast from CHC13) h (relative intensity), 504 nm (2.2), 550 nm (1.8),600 nm (LO),band edge 660nm (- 1.9 eV). GPC analysis of the THF-soluble solid polymer indicates an average molecular weight (M,)of around 10 OOO, with a polydispersity of 1.6. A CS2-soluble fraction (insoluble in CHC13 and THF) is a HTHT-coupledPHT with an identical lH NMR to the CHCl3-eoluble polymer and is thought to be a polymer of higher molecular weight. Preparation of Poly(3-octy1)thiophene Using 2 X 0.5 mol % Catalyst. The exact procedure was performed as listed in the preparation of PHTabove except on a 18mmolscale; however, after stirring for 15 h, 0.41 mol % of Ni(dppp)Clzwas added (40 mg, 0.074 mmol) at 25 "C. The solution was then stirred an additional 18h. The polymer was then precipitated with MeOH (400 mL) and allowed to sit for 2 days in MeOH, and the red precipitate was allowed to settle. The solution was decanted and the solid filtered and washed with MeOH, HzO, and MeOH again. The solid was dried under vacuum and Soxhlet extracted with MeOH and hexanes. The polymer was then dissolved in CHCl3 using a Soxhlet extractor, the CHC13 was removed, and the residue was dried and yielded 2.28 g of 96% head-to-tail coupled poly(3-octylthiophene)(65% yield). The most recent preparation gave 97 % HT-HT-coupled P O T IH NMR (CDCl3) 6.90 (8, 1H), 2.72 (t,2 H), 1.63 (m, 2 H), 1.22 (m, 10 H), 0.81 (m, 3 H), peaks also appear that represent 4% of other regiochemical isomers are assigned at 7.06 (s,0.04 H), 6.97 (s,O.001 H), 2.50 (m, 0.04 H); 13CNMR ('H decoupled) 140.0,133.8,130.6,128.5,31.9, 30.0,29.6,29.3,29.0,23.0,14.2; the relative intensity of the small absorptions (relative to the 128.5 absorption) are 140.3 (0.03), 134.8 (0.004), 127.2 (0.02), 126.7 (0.009); UV-vis (CDC13) hmax 446 nm; band edge 575 nm. (Film, cast from CHCl3) h (relative intensity) 520 (1.6), 553 (1.6),603 (l.O), band edge 675 nm (-1.8 eV). GPC analysis of the THF-soluble solid polymer indicates a average molecular weight (M,)of around 24, 424 with a polydispersity of 1.98. A Representative Quenching Reaction: Reaction of 2-Bromo-%butylthiophene with LDA, Followed by Quenching with TMSCl. 2-Bromo-3-butyl-5-(trimethylsilyl)thiophene. The quenchingstudieswere performedon a 1.5mmol was generated scale, and the 2-bromo-3-butyl-5-lithiothiophene in a procedure identical to the polymerization reactions. After an excess generation of the 2-bromo-3-butyl-5-lithiothiophene, of TMSCl was added at -60 OC and the reaction allowed to warm to room temperature overnight (-12 h). The reaction mixture was then poured onto water, the producta were extracted into (40) It is worth noting that in the case of excess n-BuLi, as expected, we have observed an increase number of nonregiospecific couplings, and in the cases of 'old" and probably wet MgBrz.OEtzand Ni(dppp)Cln,we have obtained either very low yields or no observed polymer formation.
912 J. Org. Chem., Vol. 58, No.4, 1993 EhO, dried over MgSO,, filtered, and the solvent was removed in vacuo to recover 99% of 2-bromo-3-butyl-5-(trimethylsilyl)thiophene (1HNMR 6.91 ppm, GC t~ = 12.42 min) and 1% of 2-(trimethylsilyl)-3-butyl-5-bromothiophene (6.95 ppm, GC t~ = 13.02 min) as determined by lH NMR and GUMS. Assignmenta are based on lH NMR of similar compounds prepared in our laboratory. Quenching studies on both the 5-lithium and 5-magnesium bromide of 2-bromo-3-dodecylthiophenewere performed under identical conditions. In preparation of the Grignard species, we allowed the reaction mixture containing the 5-lithiospeciesand MgBryOEhto warm to room temperature before quenching with TMSC1.
Acknowledgment. This work was partially supported by the donors of the Petroleum Research Fund, administered by the American Chemical Society, and Carnegie Mellon University 'start-up" funds. We thank the Howard
McCullough et al. Hughes Foundation for providingundergraduate research fellowshipsfor R.D.L. and D.L.A. Synthetic assistance by Shannon D. Hayes is also appreciated. We gratefully acknowledgeProf. CharlesL. Brooks, III,'and Chriatropher A. White for helpful discussions and assistance in performingthe molecular mechanicecalculationsand ab initio calculations. We also thank Shunqiong Yue and Prof. G. C. Berry for the preliminary light scattering resulta. Supplementary Material Available: lH and 13C N M R spectra of poly(3-hexylthiophene) prepared from the FeC4 method and 5a, 5c, and 5d (both N1and expanded) (14 pages). Thismaterial is contained in libraries on microfiche,immediately follows this article in the microfilm version of the journal, and can be ordered from the ACS; see any current masthead page for ordering information.
